The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted data from academic paper metadata.
Ingestion points: Reads paper metadata including titles, authors, and provenance notes from the Zotero Web API (api.zotero.org) and the local API (localhost:23119).
Boundary markers: The instructions do not specify the use of delimiters or warnings to the agent to ignore instructions embedded within paper metadata.
Capability inventory: The skill uses curl to interact with Zotero APIs, allowing it to read, create, and modify library items and tags.
Sanitization: There is no evidence of sanitization or validation of the ingested strings before they are processed by the agent, meaning a maliciously crafted paper title or note could potentially influence agent behavior.

zotero-management